INDEX

Explanations

phrases or terms associated with numerical values or codes, indicating a focus on quantitative or data-driven content

oai_token-act-pair · gpt-4o-mini Triggered by @bot

hexadecimal strings

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

indar

-0.46

 esternos

-0.45

ony

-0.45

Quin

-0.43

 Mind

-0.42

inders

-0.41

нгред

-0.41

costal

-0.40

ariya

-0.40

 passage

-0.40

POSITIVE LOGITS

fromnode

0.56

 nakalista

0.48

HasForeignKey

0.46

 sementes

0.44

 незавершена

0.42

InitStruct

0.41

 simplifié

0.41

XmlAccessorType

0.40

marle

0.37

GraphicsUnit

0.37

Activations Density 0.446%

phrases or terms associated with numerical values or codes, indicating a focus on quantitative or data-driven content

hexadecimal strings

No Comments

No Known Activations